Training in progress, step 3900, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6fa71eca77e4cb3a287115209eb353cea96380f015f1fa4a42b4b1a0444d0319
+oid sha256:85f09fb90412d0614576a7144cbf54578d067955b54bab7f83791f36182a69eb
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90ac9dd2da9c12a914155ed6d4094efe90b23009d894129ab048c74ce924d0e6
+oid sha256:b976da012ae9c6b4f84c5151834c8b18d640cbeaa9406b1c979c34864864e7d7
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cff68ccfd4b7f7441d9d6f3e2f7132d2890a46c56e8f2f0a975fd95b0cd6b66c
+oid sha256:4f387f966aecfa76017d726e43b1fa9c3ccecd8af5742b255e2ef66c7eeb645b
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3961d2f201ce99426ee113655da78d8214f6ef72fc082855daa3c565989888c2
+oid sha256:a42424f90eb68391066979b79f87ab18e2fce2a0abc640b46cf1b3f4c12c720a
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.0767734050750732,
-  "best_model_checkpoint": "./outputs/checkpoint-3800",
-  "epoch": 2.768670309653916,
+  "best_metric": 2.0676474571228027,
+  "best_model_checkpoint": "./outputs/checkpoint-3900",
+  "epoch": 2.841530054644809,
   "eval_steps": 100,
-  "global_step": 3800,
+  "global_step": 3900,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -539,13 +539,27 @@
       "eval_samples_per_second": 30.554,
       "eval_steps_per_second": 3.823,
       "step": 3800
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0002,
+      "loss": 2.0276,
+      "step": 3900
+    },
+    {
+      "epoch": 2.84,
+      "eval_loss": 2.0676474571228027,
+      "eval_runtime": 205.4014,
+      "eval_samples_per_second": 30.545,
+      "eval_steps_per_second": 3.822,
+      "step": 3900
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.0818983291332608e+17,
+  "total_flos": 1.1105217882630144e+17,
   "trial_name": null,
   "trial_params": null
 }